The goals of this project are to continue our computational and experimental investigations of HDL, focusing on mechanisms of assembly of phospholipid (PL)-rich and cholesteryl ester (CE)-rich HDL from PL-poor apoA-l. These goals are driven by our lead hypothesis that apoA-l is a uniquely elastic bilayer-bounding protein capable of absorbing PL and CE in small molecular increments. To achieve these goals, we propose, using a combination of molecular dynamics (MD) and experimental approaches, two specific aims listed in order of priority: 1) To study the structure/dynamics of PL-rich HDL. To achieve this aim, we will: i) Perform more robust simulations of the 100:2 and 50:2 particles, ii) Study the role of rotamer registry by varying registry and inhibiting buried salt bridges via MD simulations and experimentally via dimerization of Cys mutations, iii) Determine the role of the flexible domain of full length apoA-l and the effect of unesterified cholesterol (UC) and/or sphingomyelin (SM) via simulations and experimental studies, the role of salt bridge formation between lipid headgroups and apoA-l, and the mechanisms of interaction of peptide mimetics with PL-rich HDL via simulations, iv) Image PL-rich HDL assemblies using cryoEM and low angle x-ray scattering 2) To explore the molecular basis for the assembly of PL-rich HDL from PL-poor apoA-l. To achieve this aim, we will use: i) MD simulations of monomeric apoA-l from a recent x-ray crystal structure, and dimeric PLD-poor HDL using the particle shrinkage approach; ii) experimental approaches to: determine stoichiometry and kinetics of assembly of PL-rich from PL-poor HDL, determine composition/stoichiometry of pre-beta (PL-poor) HDL from plasma, and determine the structure/composition of PL-rich HDL produced by the ABCA1 pathway. Preliminary results suggest that the majority of these aims can be achieved within the 5 years of requested support. A long-term aim for which no funds are requested is: To determine the structure/dynamics of CE-rich HDL. Because of the extraordinary power of MD simulation to provide supramolecular images of HDL as indicated by our preliminary results, by combining MD simulations and experimental approaches, we are uniquely positioned to gain new insights into the structure/function of the various HDL subspecies. As a long term goal, from the resulting dynamic structure-function maps, drugs will be designed to modify the good cholesterol, HDL, to prevent or reverse coronary heart disease.